The present invention relates to a local communication system comprising a plurality of stations interconnected in a network for the communication of messages, and in particular, to the initialisation procedures of the system after start up. The invention further relates to stations for use in such a system, and methods of operation therein
A local communication system which combines source data (CD audio, MPEG video, telephone audio etc) with control messages in a low cost fibre network has been proposed in the form of D2B Optical. For details, see for example the xe2x80x9cConan Technology Brochurexe2x80x9d and the xe2x80x9cConan IC Data Sheetxe2x80x9d available from Communication and Control Electronics Limited, 2 Occam Court, Occam Road, The Surrey Research Park, Guildford, Surrey, GU2 5YQ (also http://www.candc.co.uk). See also Eurpoean patent applications of Becker GmbH EP-A-0725516 (95P03), EP-A-0725518 (95P04), EP-A-07225515 (95P05), EP-A-0725520 (95P06), EP-A-0725521 (95P07), EP-A-0725522 (95P08), EP-A-0725517 (95P09) and EP-A-0725519 (95P10). xe2x80x9cConanxe2x80x9d is a registered trade mark of Communication and Control Electronics Limited. xe2x80x9cD2Bxe2x80x9d is a registered trade mark of Philips Electronics NV.
In D2B Optical, each station obtains at an earlier stage a unique physical address, based on ring position, by a procedure described in EP-A-0725516 mentioned above, but there is still a desire for logical addressing of stations, for use at the application level. WO-A-95/01025 (PHQ 93006) describes address initialisation procedures for use with the earlier D2B network. Similar procedures can be implemented in the control message channel of the D2B Optical network.
With the system proposed and described in WO95/01025 each individual station upon start-up implements an autonomous process to find a unique device address (logical address) appropriate to its functionality. To this end, each station selects an address, and then verifies that the address is unique and available by sending a message to the address and listening for a response. In the absence of a response message the selected station assumes that the address is available and unique and adopts that as its address. If a response is received, the address is incremented and tried again until a free address is found. If no free address is found, a special default address is adopted.
According to the references, the starting point for this process is may be the last address that the station had been assigned. The hope with this type of arrangement is that the system and in particular the devices will maintain the same address from start-up to start-up. This works well once the system is stable but it has been found not so effective where the system is subject to a high degree of fluidity in its make up. A particular problem arises upon the first start up of a system after installation of several new stations. The probability increases in that case that addresses will be assigned differently in nominally identical systems, and there is even a small risk that the two stations might simultaneously acquire the same address.
The possibility with the system that two or more stations will have the same address (something that these prior patent applications are directed towards preventing) is therefore not wholly overcome. There is also a desire for a system which will address repeatably in identical installations.
The present invention aims to provide an improved system for address initialisation in networks.
In accordance with a first aspect of the present invention, there is provided a local communication system comprising a plurality of stations, one of which stations is designated a master station with the other stations designated as slave stations, which plurality of stations are interconnected in a network for the exchange of messages, wherein each of the stations has the capability to implement an autonomous process to find, and store for future use, its own unique address wherein each slave station is arranged only to begin said autonomous process to find its own address and when individually instructed by a command from the master station to do so.
In such a system the initialisation processes of the devices can be directly or indirectly controlled by the master such that two or more devices swapping addresses, or (worse) having the same address can be prevented. In particular, the master station may be arranged, at least where a group of two or more slave stations may seek addresses within the same set of addresses, to instruct the slave stations of said group one at a time. Note that the terms xe2x80x9cMasterxe2x80x9d and xe2x80x9cSlavexe2x80x9d are used in WO-A-95/01025 with a different meaning, merely indicating the source and destination respectively for a given data message. In the present case, one station is designated system master, at least for the purpose of address initialisation.
The unique address may comprise a logical address additional to a unique physical address already assigned prior to said autonomous process. The slave stations may be authorised to begin said autonomous process by means of messages addressed using said physical address. In effect, each of the stations has a multiplicity of addresses by which it can receive messages from the master and other stations on the system. In the embodiments described, these include the device address (a unique, function-related address), the broadcast address (shared with all stations), groupcast address (shared by all stations of one group) and ring position address (unique physical address). Some of these addresses may be preset in the factory whilst other of the addresses are to be set during operation of the system. In either case, stability and uniqueness of these addresses is desired.
The slave station may be provided with a physical address, and arranged to report its presence to the master station together with its physical address. The report may also specify whether the station is installed or uninstalled, and can specify the logical address already known for the installed devices. The slave may be arranged to delay sending its presence report by an amount depending on the physical address of the slave station, so that the master station will not receive presence reports from all slave stations simultaneously.
The system may be set up so that the stations are regarded as either installed or uninstalled by the master station with a new station only becoming installed when it has once been instructed by the system master to seek its own device address and has successfully achieved this. All the other stations in the system are regarded by the master station device as uninstalled, and may assume a general default device address. Installed stations may or may not also be instructed individually to seek a unique address. In the embodiment, the installed stations merely confirm that the address they have stored from a previous start-up is still unique, and only uninstalled stations require instruction from the master to seek a unique address.
Further, the system may be set up so that the installed devices are instructed by the master to initialise their addresses before the uninstalled devices are instructed to seek and initialise addresses. Alternatively, the installed devices may spontaneously act to confirm their device addresses, by a modified autonomous process.
The uninstalled devices may be instructed to seek a unique address (either strictly or approximately) in the order of their position in the network (physical address). In the embodiment described, the slave stations are arranged so as to report their uninstalled status and ring position in a staggered sequence, and the master instructs each one automatically in response to its own report. Alternatively, the master may operate by first identifying all uninstalled devices, and then instructing in turn to seek their own addresses. The master may or may not wait until each has found a unique address before instructing the next one.
In the autonomous process, the stations may seek their addresses by sending a signal to a succession of candidate addresses and monitoring responses until one is found which is free. This is similar to the verification process used in the cited earlier application.
However, in this case the devices are not all attempting to carry out the procedure at the same time so there is a significantly reduced chance of messages conflicting and not being acknowledged when they should be. Consequently, the address initialisation in the present system is more reliable in its operation.
In one particular arrangement of the system the network (optical, electrical etc) has a ring topology. In this case the above references to the position in the network are in fact references to the position in the ring.
As mentioned above, each station may begin with a general default address which it over-writes with its unique address when it has successfully found one which is available. Should the station be unable to locate an unique address it will retain the default address as its device address for the meantime.
Typically the general default address is given to all the stations in the system, at the factory. Alternatively, the stations are provided with or have a different default device address to the general default address. This may occur as a result of the station already having been used in another system or even elsewhere in the same system. In such a case, the device will typically seek to verify that the address it has is unique and if not the section may seek a unique address or ultimately revert to the general default address for the system.
An apparatus for use as a station in the present system may have volatile as well as nonvolatile memory capacity and the system may be configured so that it stores its unique address in the non-volatile memory. Installed stations can therefore remember their installed status even when power is removed from the system. xe2x80x9cStart-upxe2x80x9d as used herein may also refer to entering an operational state from a stand-by state, in which power is maintained and volatile memory may be sufficient.
When seeking a unique address the device may be configured to search for a free address within a specific range or other set of addresses. All stations may seek device addresses in the same range. As described in WO-A-95/01025, however, the range of addresses within which a device may seek to find an address may be different for different devices in the system, depending for example on device type (CD player, amplifier etc). In such a system, it is permissible for the master to instruct several stations to seek addresses at the same time, provided they are of different types.
The invention further provides an apparatus for use as a slave station with other stations in a local communication system, the apparatus comprising means for implementing after start-up an autonomous process to find and store for future use a unique address within a set of allowable addresses, wherein the apparatus when operating as a slave station is arranged to initiate said autonomous process only in response to a command sent from a master station to the slave station individually.
The invention yet further provides an apparatus for use as a master station with other stations in a local communication system, the apparatus when operating as the master station comprising means for instructing each slave station individually after start-up to implement an autonomous process to find and store for future use a unique address among a set of allowable addresses, wherein the apparatus is arranged to instruct each slave station individually to initiate said autonomous process.
The invention yet further provides a method of operating a local communication system comprising a plurality of stations of which one station is designated a master station and the other stations are designated as slave stations, plurality of stations being interconnected in a network for the exchange of messages, wherein each station has the capability autonomously to find and store for future use its own unique address within a set of allowable addresses, which method comprises the slave stations seeking a unique address only upon receipt of an instruction to do so from said master station, wherein the network includes a group of slave stations that will seek addresses in the same set of addresses, and wherein the master station instructs the slave stations of said group one at a time.